3 perldelta - what's new for perl5.004
7 This document describes differences between the 5.003 release (as
8 documented in I<Programming Perl>, second edition--the Camel Book) and
11 =head1 Supported Environments
13 Perl5.004 builds out of the box on Unix, Plan 9, LynxOS, VMS, OS/2,
14 QNX, AmigaOS, and Windows NT. Perl runs on Windows 95 as well, but it
15 cannot be built there, for lack of a reasonable command interpreter.
19 Most importantly, many bugs were fixed, including several security
20 problems. See the F<Changes> file in the distribution for details.
22 =head2 List assignment to %ENV works
24 C<%ENV = ()> and C<%ENV = @list> now work as expected (except on VMS
25 where it generates a fatal error).
27 =head2 "Can't locate Foo.pm in @INC" error now lists @INC
29 =head2 Compilation option: Binary compatibility with 5.003
31 There is a new Configure question that asks if you want to maintain
32 binary compatibility with Perl 5.003. If you choose binary
33 compatibility, you do not have to recompile your extensions, but you
34 might have symbol conflicts if you embed Perl in another application,
35 just as in the 5.003 release. By default, binary compatibility
36 is preserved at the expense of symbol table pollution.
38 =head2 $PERL5OPT environment variable
40 You may now put Perl options in the $PERL5OPT environment variable.
41 Unless Perl is running with taint checks, it will interpret this
42 variable as if its contents had appeared on a "#!perl" line at the
43 beginning of your script, except that hyphens are optional. PERL5OPT
44 may only be used to set the following switches: B<-[DIMUdmw]>.
46 =head2 Limitations on B<-M>, B<-m>, and B<-T> options
48 The C<-M> and C<-m> options are no longer allowed on the C<#!> line of
49 a script. If a script needs a module, it should invoke it with the
52 The B<-T> option is also forbidden on the C<#!> line of a script,
53 unless it was present on the Perl command line. Due to the way C<#!>
54 works, this usually means that B<-T> must be in the first argument.
59 will probably work for an executable script invoked as C<scriptname>,
64 will probably fail under the same conditions. (Non-Unix systems will
65 probably not follow this rule.) But C<perl scriptname> is guaranteed
66 to fail, since then there is no chance of B<-T> being found on the
67 command line before it is found on the C<#!> line.
69 =head2 More precise warnings
71 If you removed the B<-w> option from your Perl 5.003 scripts because it
72 made Perl too verbose, we recommend that you try putting it back when
73 you upgrade to Perl 5.004. Each new perl version tends to remove some
74 undesirable warnings, while adding new warnings that may catch bugs in
77 =head2 Deprecated: Inherited C<AUTOLOAD> for non-methods
79 Before Perl 5.004, C<AUTOLOAD> functions were looked up as methods
80 (using the C<@ISA> hierarchy), even when the function to be autoloaded
81 was called as a plain function (e.g. C<Foo::bar()>), not a method
82 (e.g. C<Foo-E<gt>bar()> or C<$obj-E<gt>bar()>).
84 Perl 5.005 will use method lookup only for methods' C<AUTOLOAD>s.
85 However, there is a significant base of existing code that may be using
86 the old behavior. So, as an interim step, Perl 5.004 issues an optional
87 warning when a non-method uses an inherited C<AUTOLOAD>.
89 The simple rule is: Inheritance will not work when autoloading
90 non-methods. The simple fix for old code is: In any module that used to
91 depend on inheriting C<AUTOLOAD> for non-methods from a base class named
92 C<BaseClass>, execute C<*AUTOLOAD = \&BaseClass::AUTOLOAD> during startup.
94 =head2 Previously deprecated %OVERLOAD is no longer usable
96 Using %OVERLOAD to define overloading was deprecated in 5.003.
97 Overloading is now defined using the overload pragma. %OVERLOAD is
98 still used internally but should not be used by Perl scripts. See
99 L<overload> for more details.
101 =head2 Subroutine arguments created only when they're modified
103 In Perl 5.004, nonexistent array and hash elements used as subroutine
104 parameters are brought into existence only if they are actually
105 assigned to (via C<@_>).
107 Earlier versions of Perl vary in their handling of such arguments.
108 Perl versions 5.002 and 5.003 always brought them into existence.
109 Perl versions 5.000 and 5.001 brought them into existence only if
110 they were not the first argument (which was almost certainly a bug).
111 Earlier versions of Perl never brought them into existence.
113 For example, given this code:
116 sub show { print $_[0] };
117 sub change { $_[0]++ };
121 After this code executes in Perl 5.004, $a{b} exists but $a[2] does
122 not. In Perl 5.002 and 5.003, both $a{b} and $a[2] would have existed
123 (but $a[2]'s value would have been undefined).
125 =head2 Group vector changeable with C<$)>
127 The C<$)> special variable has always (well, in Perl 5, at least)
128 reflected not only the current effective group, but also the group list
129 as returned by the C<getgroups()> C function (if there is one).
130 However, until this release, there has not been a way to call the
131 C<setgroups()> C function from Perl.
133 In Perl 5.004, assigning to C<$)> is exactly symmetrical with examining
134 it: The first number in its string value is used as the effective gid;
135 if there are any numbers after the first one, they are passed to the
136 C<setgroups()> C function (if there is one).
138 =head2 Fixed parsing of $$<digit>, &$<digit>, etc.
140 Perl versions before 5.004 misinterpreted any type marker followed by
141 "$" and a digit. For example, "$$0" was incorrectly taken to mean
142 "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004.
144 However, the developers of Perl 5.004 could not fix this bug completely,
145 because at least two widely-used modules depend on the old meaning of
146 "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the
147 old (broken) way inside strings; but it generates this message as a
148 warning. And in Perl 5.005, this special treatment will cease.
150 =head2 No resetting of $. on implicit close
152 The documentation for Perl 5.0 has always stated that C<$.> is I<not>
153 reset when an already-open file handle is reopened with no intervening
154 call to C<close>. Due to a bug, perl versions 5.000 through 5.003
155 I<did> reset C<$.> under that circumstance; Perl 5.004 does not.
157 =head2 C<wantarray> may return undef
159 The C<wantarray> operator returns true if a subroutine is expected to
160 return a list, and false otherwise. In Perl 5.004, C<wantarray> can
161 also return the undefined value if a subroutine's return value will
162 not be used at all, which allows subroutines to avoid a time-consuming
163 calculation of a return value if it isn't going to be used.
165 =head2 Changes to tainting checks
167 A bug in previous versions may have failed to detect some insecure
168 conditions when taint checks are turned on. (Taint checks are used
169 in setuid or setgid scripts, or when explicitly turned on with the
170 C<-T> invocation option.) Although it's unlikely, this may cause a
171 previously-working script to now fail -- which should be construed
172 as a blessing, since that indicates a potentially-serious security
173 hole was just plugged.
175 The new restrictions when tainting include:
179 =item No glob() or <*>
181 These operators may spawn the C shell (csh), which cannot be made
182 safe. This restriction will be lifted in a future version of Perl
183 when globbing is implemented without the use of an external program.
185 =item No spawning if tainted $CDPATH, $ENV, $BASH_ENV
187 These environment variables may alter the behavior of spawned programs
188 (especially shells) in ways that subvert security. So now they are
189 treated as dangerous, in the manner of $IFS and $PATH.
191 =item No spawning if tainted $TERM doesn't look like a terminal name
193 Some termcap libraries do unsafe things with $TERM. However, it would be
194 unnecessarily harsh to treat all $TERM values as unsafe, since only shell
195 metacharacters can cause trouble in $TERM. So a tainted $TERM is
196 considered to be safe if it contains only alphanumerics, underscores,
197 dashes, and colons, and unsafe if it contains other characters (including
202 =head2 New Opcode module and revised Safe module
204 A new Opcode module supports the creation, manipulation and
205 application of opcode masks. The revised Safe module has a new API
206 and is implemented using the new Opcode module. Please read the new
207 Opcode and Safe documentation.
209 =head2 Embedding improvements
211 In older versions of Perl it was not possible to create more than one
212 Perl interpreter instance inside a single process without leaking like a
213 sieve and/or crashing. The bugs that caused this behavior have all been
214 fixed. However, you still must take care when embedding Perl in a C
215 program. See the updated perlembed manpage for tips on how to manage
218 =head2 Internal change: FileHandle class based on IO::* classes
220 File handles are now stored internally as type IO::Handle. The
221 FileHandle module is still supported for backwards compatibility, but
222 it is now merely a front end to the IO::* modules -- specifically,
223 IO::Handle, IO::Seekable, and IO::File. We suggest, but do not
224 require, that you use the IO::* modules in new code.
226 In harmony with this change, C<*GLOB{FILEHANDLE}> is now just a
227 backward-compatible synonym for C<*GLOB{IO}>.
229 =head2 Internal change: PerlIO abstraction interface
231 It is now possible to build Perl with AT&T's sfio IO package
232 instead of stdio. See L<perlapio> for more details, and
233 the F<INSTALL> file for how to use it.
235 =head2 New and changed syntax
239 =item $coderef->(PARAMS)
241 A subroutine reference may now be suffixed with an arrow and a
242 (possibly empty) parameter list. This syntax denotes a call of the
243 referenced subroutine, with the given parameters (if any).
245 This new syntax follows the pattern of S<C<$hashref-E<gt>{FOO}>> and
246 S<C<$aryref-E<gt>[$foo]>>: You may now write S<C<&$subref($foo)>> as
247 S<C<$subref-E<gt>($foo)>>. All of these arrow terms may be chained;
248 thus, S<C<&{$table-E<gt>{FOO}}($bar)>> may now be written
249 S<C<$table-E<gt>{FOO}-E<gt>($bar)>>.
253 =head2 New and changed builtin constants
259 The current package name at compile time, or the undefined value if
260 there is no current package (due to a C<package;> directive). Like
261 C<__FILE__> and C<__LINE__>, C<__PACKAGE__> does I<not> interpolate
266 =head2 New and changed builtin variables
272 Extended error message on some platforms. (Also known as
273 $EXTENDED_OS_ERROR if you C<use English>).
277 The current set of syntax checks enabled by C<use strict>. See the
278 documentation of C<strict> for more details. Not actually new, but
280 Because it is intended for internal use by Perl core components,
281 there is no C<use English> long name for this variable.
285 By default, running out of memory it is not trappable. However, if
286 compiled for this, Perl may use the contents of C<$^M> as an emergency
287 pool after die()ing with this message. Suppose that your Perl were
288 compiled with -DEMERGENCY_SBRK and used Perl's malloc. Then
292 would allocate a 64K buffer for use when in emergency.
293 See the F<INSTALL> file for information on how to enable this option.
294 As a disincentive to casual use of this advanced feature,
295 there is no C<use English> long name for this variable.
299 =head2 New and changed builtin functions
303 =item delete on slices
305 This now works. (e.g. C<delete @ENV{'PATH', 'MANPATH'}>)
309 is now supported on more platforms, prefers fcntl to lockf when
310 emulating, and always flushes before (un)locking.
312 =item printf and sprintf
314 Perl now implements these functions itself; it doesn't use the C
315 library function sprintf() any more, except for floating-point
316 numbers, and even then only known flags are allowed. As a result, it
317 is now possible to know which conversions and flags will work, and
320 The new conversions in Perl's sprintf() are:
323 %p a pointer (the address of the Perl value, in hexadecimal)
324 %n special: *stores* the number of characters output so far
325 into the next variable in the parameter list
327 The new flags that go between the C<%> and the conversion are:
329 # prefix octal with "0", hex with "0x"
330 h interpret integer as C type "short" or "unsigned short"
331 V interpret integer as Perl's standard integer type
333 Also, where a number would appear in the flags, an asterisk ("*") may
334 be used instead, in which case Perl uses the next item in the
335 parameter list as the given number (that is, as the field width or
336 precision). If a field width obtained through "*" is negative, it has
337 the same effect as the '-' flag: left-justification.
339 See L<perlfunc/sprintf> for a complete list of conversion and flags.
341 =item keys as an lvalue
343 As an lvalue, C<keys> allows you to increase the number of hash buckets
344 allocated for the given hash. This can gain you a measure of efficiency if
345 you know the hash is going to get big. (This is similar to pre-extending
346 an array by assigning a larger number to $#array.) If you say
350 then C<%hash> will have at least 200 buckets allocated for it. These
351 buckets will be retained even if you do C<%hash = ()>; use C<undef
352 %hash> if you want to free the storage while C<%hash> is still in scope.
353 You can't shrink the number of buckets allocated for the hash using
354 C<keys> in this way (but you needn't worry about doing this by accident,
355 as trying has no effect).
357 =item my() in Control Structures
359 You can now use my() (with or without the parentheses) in the control
360 expressions of control structures such as:
362 while (defined(my $line = <>)) {
368 if ((my $answer = <STDIN>) =~ /^y(es)?$/i) {
370 } elsif ($answer =~ /^n(o)?$/i) {
374 die "`$answer' is neither `yes' nor `no'";
377 Also, you can declare a foreach loop control variable as lexical by
378 preceding it with the word "my". For example, in:
380 foreach my $i (1, 2, 3) {
384 $i is a lexical variable, and the scope of $i extends to the end of
385 the loop, but not beyond it.
387 Note that you still cannot use my() on global punctuation variables
388 such as $_ and the like.
390 =item pack() and unpack()
392 A new format 'w' represents a BER compressed integer (as defined in
393 ASN.1). Its format is a sequence of one or more bytes, each of which
394 provides seven bits of the total value, with the most significant
395 first. Bit eight of each byte is set, except for the last byte, in
396 which bit eight is clear.
398 Both pack() and unpack() now fail when their templates contain invalid
399 types. (Invalid types used to be ignored.)
403 The new sysseek() operator is a variant of seek() that sets and gets the
404 file's system read/write position, using the lseek(2) system call. It is
405 the only reliable way to seek before using sysread() or syswrite(). Its
406 return value is the new position, or the undefined value on failure.
410 If the first argument to C<use> is a number, it is treated as a version
411 number instead of a module name. If the version of the Perl interpreter
412 is less than VERSION, then an error message is printed and Perl exits
413 immediately. Because C<use> occurs at compile time, this check happens
414 immediately during the compilation process, unlike C<require VERSION>,
415 which waits until runtime for the check. This is often useful if you
416 need to check the current Perl version before C<use>ing library modules
417 which have changed in incompatible ways from older versions of Perl.
418 (We try not to do this more than we have to.)
420 =item use Module VERSION LIST
422 If the VERSION argument is present between Module and LIST, then the
423 C<use> will call the VERSION method in class Module with the given
424 version as an argument. The default VERSION method, inherited from
425 the UNIVERSAL class, croaks if the given version is larger than the
426 value of the variable $Module::VERSION. (Note that there is not a
427 comma after VERSION!)
429 This version-checking mechanism is similar to the one currently used
430 in the Exporter module, but it is faster and can be used with modules
431 that don't use the Exporter. It is the recommended method for new
434 =item prototype(FUNCTION)
436 Returns the prototype of a function as a string (or C<undef> if the
437 function has no prototype). FUNCTION is a reference to or the name of the
438 function whose prototype you want to retrieve.
439 (Not actually new; just never documented before.)
443 The default seed for C<srand>, which used to be C<time>, has been changed.
444 Now it's a heady mix of difficult-to-predict system-dependent values,
445 which should be sufficient for most everyday purposes.
447 Previous to version 5.004, calling C<rand> without first calling C<srand>
448 would yield the same sequence of random numbers on most or all machines.
449 Now, when perl sees that you're calling C<rand> and haven't yet called
450 C<srand>, it calls C<srand> with the default seed. You should still call
451 C<srand> manually if your code might ever be run on a pre-5.004 system,
452 of course, or if you want a seed other than the default.
456 Functions documented in the Camel to default to $_ now in
457 fact do, and all those that do are so documented in L<perlfunc>.
459 =item C<m//gc> does not reset search position on failure
461 The C<m//g> match iteration construct has always reset its target
462 string's search position (which is visible through the C<pos> operator)
463 when a match fails; as a result, the next C<m//g> match after a failure
464 starts again at the beginning of the string. With Perl 5.004, this
465 reset may be disabled by adding the "c" (for "continue") modifier,
466 i.e. C<m//gc>. This feature, in conjunction with the C<\G> zero-width
467 assertion, makes it possible to chain matches together. See L<perlop>
470 =item C<m//x> ignores whitespace before ?*+{}
472 The C<m//x> construct has always been intended to ignore all unescaped
473 whitespace. However, before Perl 5.004, whitespace had the effect of
474 escaping repeat modifiers like "*" or "?"; for example, C</a *b/x> was
475 (mis)interpreted as C</a\*b/x>. This bug has been fixed in 5.004.
477 =item nested C<sub{}> closures work now
479 Prior to the 5.004 release, nested anonymous functions didn't work
482 =item formats work right on changing lexicals
484 Just like anonymous functions that contain lexical variables
485 that change (like a lexical index variable for a C<foreach> loop),
486 formats now work properly. For example, this silently failed
487 before (printed only zeros), but is fine now:
490 foreach $i ( 1 .. 10 ) {
500 =head2 New builtin methods
502 The C<UNIVERSAL> package automatically contains the following methods that
503 are inherited by all other classes:
509 C<isa> returns I<true> if its object is blessed into a subclass of C<CLASS>
511 C<isa> is also exportable and can be called as a sub with two arguments. This
512 allows the ability to check what a reference points to. Example:
514 use UNIVERSAL qw(isa);
516 if(isa($ref, 'ARRAY')) {
522 C<can> checks to see if its object has a method called C<METHOD>,
523 if it does then a reference to the sub is returned; if it does not then
524 I<undef> is returned.
526 =item VERSION( [NEED] )
528 C<VERSION> returns the version number of the class (package). If the
529 NEED argument is given then it will check that the current version (as
530 defined by the $VERSION variable in the given package) not less than
531 NEED; it will die if this is not the case. This method is normally
532 called as a class method. This method is called automatically by the
533 C<VERSION> form of C<use>.
535 use A 1.2 qw(some imported subs);
541 B<NOTE:> C<can> directly uses Perl's internal code for method lookup, and
542 C<isa> uses a very similar method and caching strategy. This may cause
543 strange effects if the Perl code dynamically changes @ISA in any package.
545 You may add other methods to the UNIVERSAL class via Perl or XS code.
546 You do not need to C<use UNIVERSAL> in order to make these methods
547 available to your program. This is necessary only if you wish to
548 have C<isa> available as a plain subroutine in the current package.
550 =head2 TIEHANDLE now supported
552 See L<perltie> for other kinds of tie()s.
556 =item TIEHANDLE classname, LIST
558 This is the constructor for the class. That means it is expected to
559 return an object of some sort. The reference can be used to
560 hold some internal information.
565 return bless \$i, shift;
568 =item PRINT this, LIST
570 This method will be triggered every time the tied handle is printed to.
571 Beyond its self reference it also expects the list that was passed to
577 return print join( $, => map {uc} @_), $\;
580 =item PRINTF this, LIST
582 This method will be triggered every time the tied handle is printed to
583 with the C<printf()> function.
584 Beyond its self reference it also expects the format and list that was
585 passed to the printf function.
590 print sprintf($fmt, @_)."\n";
595 This method will be called when the handle is read from via the C<read>
596 or C<sysread> functions.
600 my($buf,$len,$offset) = @_;
601 print "READ called, \$buf=$buf, \$len=$len, \$offset=$offset";
606 This method will be called when the handle is read from. The method
607 should return undef when there is no more data.
611 return "PRINT called $$r times\n"
616 This method will be called when the C<getc> function is called.
618 sub GETC { print "Don't GETC, Get Perl"; return "a"; }
622 As with the other types of ties, this method will be called when the
623 tied handle is about to be destroyed. This is useful for debugging and
624 possibly for cleaning up.
632 =head2 Malloc enhancements
634 Four new compilation flags are recognized by malloc.c. (They have no
635 effect if perl is compiled with system malloc().)
639 =item -DDEBUGGING_MSTATS
641 If perl is compiled with C<DEBUGGING_MSTATS> defined, you can print
642 memory statistics at runtime by running Perl thusly:
644 env PERL_DEBUG_MSTATS=2 perl your_script_here
646 The value of 2 means to print statistics after compilation and on
647 exit; with a value of 1, the statistics ares printed only on exit.
648 (If you want the statistics at an arbitrary time, you'll need to
649 install the optional module Devel::Peek.)
651 =item -DEMERGENCY_SBRK
653 If this macro is defined, running out of memory need not be a fatal
654 error: a memory pool can allocated by assigning to the special
655 variable C<$^M>. See L<"$^M">.
659 Perl memory allocation is by bucket with sizes close to powers of two.
660 Because of these malloc overhead may be big, especially for data of
661 size exactly a power of two. If C<PACK_MALLOC> is defined, perl uses
662 a slightly different algorithm for small allocations (up to 64 bytes
663 long), which makes it possible to have overhead down to 1 byte for
664 allocations which are powers of two (and appear quite often).
666 Expected memory savings (with 8-byte alignment in C<alignbytes>) is
667 about 20% for typical Perl usage. Expected slowdown due to additional
668 malloc overhead is in fractions of a percent (hard to measure, because
669 of the effect of saved memory on speed).
671 =item -DTWO_POT_OPTIMIZE
673 Similarly to C<PACK_MALLOC>, this macro improves allocations of data
674 with size close to a power of two; but this works for big allocations
675 (starting with 16K by default). Such allocations are typical for big
676 hashes and special-purpose scripts, especially image processing.
678 On recent systems, the fact that perl requires 2M from system for 1M
679 allocation will not affect speed of execution, since the tail of such
680 a chunk is not going to be touched (and thus will not require real
681 memory). However, it may result in a premature out-of-memory error.
682 So if you will be manipulating very large blocks with sizes close to
683 powers of two, it would be wise to define this macro.
685 Expected saving of memory is 0-100% (100% in applications which
686 require most memory in such 2**n chunks); expected slowdown is
691 =head2 Miscellaneous efficiency enhancements
693 Functions that have an empty prototype and that do nothing but return
694 a fixed value are now inlined (e.g. C<sub PI () { 3.14159 }>).
696 Each unique hash key is only allocated once, no matter how many hashes
697 have an entry with that key. So even if you have 100 copies of the
698 same hash, the hash keys never have to be reallocated.
700 =head1 Support for More Operating Systems
702 Support for the following operating systems is new in Perl 5.004.
706 Perl 5.004 now includes support for building a "native" perl under
707 Windows NT, using the Microsoft Visual C++ compiler (versions 2.0
708 and above) or the Borland C++ compiler (versions 5.02 and above).
709 The resulting perl can be used under Windows 95 (if it
710 is installed in the same directory locations as it got installed
711 in Windows NT). This port includes support for perl extension
712 building tools like L<MakeMaker> and L<h2xs>, so that many extensions
713 available on the Comprehensive Perl Archive Network (CPAN) can now be
714 readily built under Windows NT. See http://www.perl.com/ for more
715 information on CPAN, and L<README.win32> for more details on how to
716 get started with building this port.
718 There is also support for building perl under the Cygwin32 environment.
719 Cygwin32 is a set of GNU tools that make it possible to compile and run
720 many UNIX programs under Windows NT by providing a mostly UNIX-like
721 interface for compilation and execution. See L<README.cygwin32> for
722 more details on this port, and how to obtain the Cygwin32 toolkit.
734 See L<README.amigaos>.
738 Six new pragmatic modules exist:
742 =item use autouse MODULE => qw(sub1 sub2 sub3)
744 Defers C<require MODULE> until someone calls one of the specified
745 subroutines (which must be exported by MODULE). This pragma should be
746 used with caution, and only when necessary.
752 Looks for MakeMaker-like I<'blib'> directory structure starting in
753 I<dir> (or current directory) and working back up to five levels of
756 Intended for use on command line with B<-M> option as a way of testing
757 arbitrary scripts against an uninstalled version of a package.
759 =item use constant NAME => VALUE
761 Provides a convenient interface for creating compile-time constants,
762 See L<perlsub/"Constant Functions">.
766 Tells the compiler to enable (or disable) the use of POSIX locales for
769 When C<use locale> is in effect, the current LC_CTYPE locale is used
770 for regular expressions and case mapping; LC_COLLATE for string
771 ordering; and LC_NUMERIC for numeric formating in printf and sprintf
772 (but B<not> in print). LC_NUMERIC is always used in write, since
773 lexical scoping of formats is problematic at best.
775 Each C<use locale> or C<no locale> affects statements to the end of
776 the enclosing BLOCK or, if not inside a BLOCK, to the end of the
777 current file. Locales can be switched and queried with
780 See L<perllocale> for more information.
784 Disable unsafe opcodes, or any named opcodes, when compiling Perl code.
788 Enable VMS-specific language features. Currently, there are three
789 VMS-specific features available: 'status', which makes C<$?> and
790 C<system> return genuine VMS status values instead of emulating POSIX;
791 'exit', which makes C<exit> take a genuine VMS status value instead of
792 assuming that C<exit 1> is an error; and 'time', which makes all times
793 relative to the local time zone, in the VMS tradition.
799 =head2 Required Updates
801 Though Perl 5.004 is compatible with almost all modules that work
802 with Perl 5.003, there are a few exceptions:
804 Module Required Version for Perl 5.004
805 ------ -------------------------------
808 Tk Tk400.202 (-w makes noise)
810 Also, the majordomo mailing list program, version 1.94.1, doesn't work
811 with Perl 5.004 (nor with perl 4), because it executes an invalid
812 regular expression. This bug is fixed in majordomo version 1.94.2.
814 =head2 Installation directories
816 The I<installperl> script now places the Perl source files for
817 extensions in the architecture-specific library directory, which is
818 where the shared libraries for extensions have always been. This
819 change is intended to allow administrators to keep the Perl 5.004
820 library directory unchanged from a previous version, without running
821 the risk of binary incompatibility between extensions' Perl source and
824 =head2 Module information summary
826 Brand new modules, arranged by topic rather than strictly
829 CGI.pm Web server interface ("Common Gateway Interface")
830 CGI/Apache.pm Support for Apache's Perl module
831 CGI/Carp.pm Log server errors with helpful context
832 CGI/Fast.pm Support for FastCGI (persistent server process)
833 CGI/Push.pm Support for server push
834 CGI/Switch.pm Simple interface for multiple server types
836 CPAN Interface to Comprehensive Perl Archive Network
837 CPAN::FirstTime Utility for creating CPAN configuration file
838 CPAN::Nox Runs CPAN while avoiding compiled extensions
840 IO.pm Top-level interface to IO::* classes
841 IO/File.pm IO::File extension Perl module
842 IO/Handle.pm IO::Handle extension Perl module
843 IO/Pipe.pm IO::Pipe extension Perl module
844 IO/Seekable.pm IO::Seekable extension Perl module
845 IO/Select.pm IO::Select extension Perl module
846 IO/Socket.pm IO::Socket extension Perl module
848 Opcode.pm Disable named opcodes when compiling Perl code
850 ExtUtils/Embed.pm Utilities for embedding Perl in C programs
851 ExtUtils/testlib.pm Fixes up @INC to use just-built extension
853 FindBin.pm Find path of currently executing program
855 Class/Struct.pm Declare struct-like datatypes as Perl classes
856 File/stat.pm By-name interface to Perl's builtin stat
857 Net/hostent.pm By-name interface to Perl's builtin gethost*
858 Net/netent.pm By-name interface to Perl's builtin getnet*
859 Net/protoent.pm By-name interface to Perl's builtin getproto*
860 Net/servent.pm By-name interface to Perl's builtin getserv*
861 Time/gmtime.pm By-name interface to Perl's builtin gmtime
862 Time/localtime.pm By-name interface to Perl's builtin localtime
863 Time/tm.pm Internal object for Time::{gm,local}time
864 User/grent.pm By-name interface to Perl's builtin getgr*
865 User/pwent.pm By-name interface to Perl's builtin getpw*
867 Tie/RefHash.pm Base class for tied hashes with references as keys
869 UNIVERSAL.pm Base class for *ALL* classes
873 New constants in the existing Fcntl modules are now supported,
874 provided that your operating system happens to support them:
877 O_ASYNC O_DEFER O_DSYNC O_FSYNC O_SYNC
880 These constants are intended for use with the Perl operators sysopen()
881 and fcntl() and the basic database modules like SDBM_File. For the
882 exact meaning of these and other Fcntl constants please refer to your
883 operating system's documentation for fcntl() and open().
885 In addition, the Fcntl module now provides these constants for use
886 with the Perl operator flock():
888 LOCK_SH LOCK_EX LOCK_NB LOCK_UN
890 These constants are defined in all environments (because where there is
891 no flock() system call, Perl emulates it). However, for historical
892 reasons, these constants are not exported unless they are explicitly
893 requested with the ":flock" tag (e.g. C<use Fcntl ':flock'>).
897 The IO module provides a simple mechanism to load all of the IO modules at one
898 go. Currently this includes:
906 For more information on any of these modules, please see its
907 respective documentation.
911 The Math::Complex module has been totally rewritten, and now supports
912 more operations. These are overloaded:
914 + - * / ** <=> neg ~ abs sqrt exp log sin cos atan2 "" (stringify)
916 And these functions are now exported:
919 log10 logn ln cbrt root
932 This new module provides a simpler interface to parts of Math::Complex for
933 those who need trigonometric functions only for real numbers.
937 There have been quite a few changes made to DB_File. Here are a few of
944 Fixed a handful of bugs.
948 By public demand, added support for the standard hash function exists().
952 Made it compatible with Berkeley DB 1.86.
956 Made negative subscripts work with RECNO interface.
960 Changed the default flags from O_RDWR to O_CREAT|O_RDWR and the default
961 mode from 0640 to 0666.
965 Made DB_File automatically import the open() constants (O_RDWR,
966 O_CREAT etc.) from Fcntl, if available.
970 Updated documentation.
974 Refer to the HISTORY section in DB_File.pm for a complete list of
975 changes. Everything after DB_File 1.01 has been added since 5.003.
979 Major rewrite - support added for both udp echo and real icmp pings.
981 =head2 Object-oriented overrides for builtin operators
983 Many of the Perl builtins returning lists now have
984 object-oriented overrides. These are:
996 For example, you can now say
1000 $his = (stat($filename)->st_uid == pwent($whoever)->pw_uid);
1002 =head1 Utility Changes
1008 =item Sends converted HTML to standard output
1010 The I<pod2html> utility included with Perl 5.004 is entirely new.
1011 By default, it sends the converted HTML to its standard output,
1012 instead of writing it to a file like Perl 5.003's I<pod2html> did.
1013 Use the B<--outfile=FILENAME> option to write to a file.
1021 =item C<void> XSUBs now default to returning nothing
1023 Due to a documentation/implementation bug in previous versions of
1024 Perl, XSUBs with a return type of C<void> have actually been
1025 returning one value. Usually that value was the GV for the XSUB,
1026 but sometimes it was some already freed or reused value, which would
1027 sometimes lead to program failure.
1029 In Perl 5.004, if an XSUB is declared as returning C<void>, it
1030 actually returns no value, i.e. an empty list (though there is a
1031 backward-compatibility exception; see below). If your XSUB really
1032 does return an SV, you should give it a return type of C<SV *>.
1034 For backward compatibility, I<xsubpp> tries to guess whether a
1035 C<void> XSUB is really C<void> or if it wants to return an C<SV *>.
1036 It does so by examining the text of the XSUB: if I<xsubpp> finds
1037 what looks like an assignment to C<ST(0)>, it assumes that the
1038 XSUB's return type is really C<SV *>.
1042 =head1 C Language API Changes
1046 =item C<gv_fetchmethod> and C<perl_call_sv>
1048 The C<gv_fetchmethod> function finds a method for an object, just like
1049 in Perl 5.003. The GV it returns may be a method cache entry.
1050 However, in Perl 5.004, method cache entries are not visible to users;
1051 therefore, they can no longer be passed directly to C<perl_call_sv>.
1052 Instead, you should use the C<GvCV> macro on the GV to extract its CV,
1053 and pass the CV to C<perl_call_sv>.
1055 The most likely symptom of passing the result of C<gv_fetchmethod> to
1056 C<perl_call_sv> is Perl's producing an "Undefined subroutine called"
1057 error on the I<second> call to a given method (since there is no cache
1060 =item C<perl_eval_pv>
1062 A new function handy for eval'ing strings of Perl code inside C code.
1063 This function returns the value from the eval statement, which can
1064 be used instead of fetching globals from the symbol table. See
1065 L<perlguts>, L<perlembed> and L<perlcall> for details and examples.
1067 =item Extended API for manipulating hashes
1069 Internal handling of hash keys has changed. The old hashtable API is
1070 still fully supported, and will likely remain so. The additions to the
1071 API allow passing keys as C<SV*>s, so that C<tied> hashes can be given
1072 real scalars as keys rather than plain strings (nontied hashes still
1073 can only use strings as keys). New extensions must use the new hash
1074 access functions and macros if they wish to use C<SV*> keys. These
1075 additions also make it feasible to manipulate C<HE*>s (hash entries),
1076 which can be more efficient. See L<perlguts> for details.
1080 =head1 Documentation Changes
1082 Many of the base and library pods were updated. These
1083 new pods are included in section 1:
1093 Frequently asked questions.
1097 Locale support (internationalization and localization).
1101 Tutorial on Perl OO programming.
1105 Perl internal IO abstraction interface.
1109 Perl module library and recommended practice for module creation.
1110 Extracted from L<perlmod> (which is much smaller as a result).
1114 Although not new, this has been massively updated.
1118 Although not new, this has been massively updated.
1122 =head1 New Diagnostics
1124 Several new conditions will trigger warnings that were
1125 silent before. Some only affect certain platforms.
1126 The following new warnings and errors outline these.
1127 These messages are classified as follows (listed in
1128 increasing order of desperation):
1130 (W) A warning (optional).
1131 (D) A deprecation (optional).
1132 (S) A severe warning (mandatory).
1133 (F) A fatal error (trappable).
1134 (P) An internal error you should never see (trappable).
1135 (X) A very fatal error (nontrappable).
1136 (A) An alien error message (not generated by Perl).
1140 =item "my" variable %s masks earlier declaration in same scope
1142 (S) A lexical variable has been redeclared in the same scope, effectively
1143 eliminating all access to the previous instance. This is almost always
1144 a typographical error. Note that the earlier variable will still exist
1145 until the end of the scope or until all closure referents to it are
1148 =item %s argument is not a HASH element or slice
1150 (F) The argument to delete() must be either a hash element, such as
1153 $ref->[12]->{"susie"}
1155 or a hash slice, such as
1157 @foo{$bar, $baz, $xyzzy}
1158 @{$ref->[12]}{"susie", "queue"}
1160 =item Allocation too large: %lx
1162 (X) You can't allocate more than 64K on an MS-DOS machine.
1164 =item Allocation too large
1166 (F) You can't allocate more than 2^31+"small amount" bytes.
1168 =item Applying %s to %s will act on scalar(%s)
1170 (W) The pattern match (//), substitution (s///), and translation (tr///)
1171 operators work on scalar values. If you apply one of them to an array
1172 or a hash, it will convert the array or hash to a scalar value -- the
1173 length of an array, or the population info of a hash -- and then work on
1174 that scalar value. This is probably not what you meant to do. See
1175 L<perlfunc/grep> and L<perlfunc/map> for alternatives.
1177 =item Attempt to free nonexistent shared string
1179 (P) Perl maintains a reference counted internal table of strings to
1180 optimize the storage and access of hash keys and other strings. This
1181 indicates someone tried to decrement the reference count of a string
1182 that can no longer be found in the table.
1184 =item Attempt to use reference as lvalue in substr
1186 (W) You supplied a reference as the first argument to substr() used
1187 as an lvalue, which is pretty strange. Perhaps you forgot to
1188 dereference it first. See L<perlfunc/substr>.
1190 =item Can't redefine active sort subroutine %s
1192 (F) Perl optimizes the internal handling of sort subroutines and keeps
1193 pointers into them. You tried to redefine one such sort subroutine when it
1194 was currently active, which is not allowed. If you really want to do
1195 this, you should write C<sort { &func } @x> instead of C<sort func @x>.
1197 =item Can't use bareword ("%s") as %s ref while "strict refs" in use
1199 (F) Only hard references are allowed by "strict refs". Symbolic references
1200 are disallowed. See L<perlref>.
1202 =item Cannot resolve method `%s' overloading `%s' in package `%s'
1204 (P) Internal error trying to resolve overloading specified by a method
1205 name (as opposed to a subroutine reference).
1207 =item Constant subroutine %s redefined
1209 (S) You redefined a subroutine which had previously been eligible for
1210 inlining. See L<perlsub/"Constant Functions"> for commentary and
1213 =item Constant subroutine %s undefined
1215 (S) You undefined a subroutine which had previously been eligible for
1216 inlining. See L<perlsub/"Constant Functions"> for commentary and
1219 =item Copy method did not return a reference
1221 (F) The method which overloads "=" is buggy. See L<overload/Copy Constructor>.
1225 (F) You passed die() an empty string (the equivalent of C<die "">) or
1226 you called it with no args and both C<$@> and C<$_> were empty.
1228 =item Exiting pseudo-block via %s
1230 (W) You are exiting a rather special block construct (like a sort block or
1231 subroutine) by unconventional means, such as a goto, or a loop control
1232 statement. See L<perlfunc/sort>.
1234 =item Identifier too long
1236 (F) Perl limits identifiers (names for variables, functions, etc.) to
1237 252 characters for simple names, somewhat more for compound names (like
1238 C<$A::B>). You've exceeded Perl's limits. Future versions of Perl are
1239 likely to eliminate these arbitrary limitations.
1241 =item Illegal character %s (carriage return)
1243 (F) A carriage return character was found in the input. This is an
1244 error, and not a warning, because carriage return characters can break
1245 multi-line strings, including here documents (e.g., C<print E<lt>E<lt>EOF;>).
1247 =item Illegal switch in PERL5OPT: %s
1249 (X) The PERL5OPT environment variable may only be used to set the
1250 following switches: B<-[DIMUdmw]>.
1252 =item Integer overflow in hex number
1254 (S) The literal hex number you have specified is too big for your
1255 architecture. On a 32-bit architecture the largest hex literal is
1258 =item Integer overflow in octal number
1260 (S) The literal octal number you have specified is too big for your
1261 architecture. On a 32-bit architecture the largest octal literal is
1264 =item internal error: glob failed
1266 (P) Something went wrong with the external program(s) used for C<glob>
1267 and C<E<lt>*.cE<gt>>. This may mean that your csh (C shell) is
1268 broken. If so, you should change all of the csh-related variables in
1269 config.sh: If you have tcsh, make the variables refer to it as if it
1270 were csh (e.g. C<full_csh='/usr/bin/tcsh'>); otherwise, make them all
1271 empty (except that C<d_csh> should be C<'undef'>) so that Perl will
1272 think csh is missing. In either case, after editing config.sh, run
1273 C<./Configure -S> and rebuild Perl.
1275 =item Invalid conversion in %s: "%s"
1277 (W) Perl does not understand the given format conversion.
1278 See L<perlfunc/sprintf>.
1280 =item Invalid type in pack: '%s'
1282 (F) The given character is not a valid pack type. See L<perlfunc/pack>.
1284 =item Invalid type in unpack: '%s'
1286 (F) The given character is not a valid unpack type. See L<perlfunc/unpack>.
1288 =item Name "%s::%s" used only once: possible typo
1290 (W) Typographical errors often show up as unique variable names.
1291 If you had a good reason for having a unique name, then just mention
1292 it again somehow to suppress the message (the C<use vars> pragma is
1293 provided for just this purpose).
1295 =item Null picture in formline
1297 (F) The first argument to formline must be a valid format picture
1298 specification. It was found to be empty, which probably means you
1299 supplied it an uninitialized value. See L<perlform>.
1301 =item Offset outside string
1303 (F) You tried to do a read/write/send/recv operation with an offset
1304 pointing outside the buffer. This is difficult to imagine.
1305 The sole exception to this is that C<sysread()>ing past the buffer
1306 will extend the buffer and zero pad the new area.
1308 =item Out of memory!
1310 (X|F) The malloc() function returned 0, indicating there was insufficient
1311 remaining memory (or virtual memory) to satisfy the request.
1313 The request was judged to be small, so the possibility to trap it
1314 depends on the way Perl was compiled. By default it is not trappable.
1315 However, if compiled for this, Perl may use the contents of C<$^M> as
1316 an emergency pool after die()ing with this message. In this case the
1317 error is trappable I<once>.
1319 =item Out of memory during request for %s
1321 (F) The malloc() function returned 0, indicating there was insufficient
1322 remaining memory (or virtual memory) to satisfy the request. However,
1323 the request was judged large enough (compile-time default is 64K), so
1324 a possibility to shut down by trapping this error is granted.
1328 (P) The library function frexp() failed, making printf("%f") impossible.
1330 =item Possible attempt to put comments in qw() list
1332 (W) qw() lists contain items separated by whitespace; as with literal
1333 strings, comment characters are not ignored, but are instead treated
1334 as literal data. (You may have used different delimiters than the
1335 exclamation marks parentheses shown here; braces are also frequently
1338 You probably wrote something like this:
1345 when you should have written this:
1352 If you really want comments, build your list the
1353 old-fashioned way, with quotes and commas:
1357 'b', # another comment
1360 =item Possible attempt to separate words with commas
1362 (W) qw() lists contain items separated by whitespace; therefore commas
1363 aren't needed to separate the items. (You may have used different
1364 delimiters than the parentheses shown here; braces are also frequently
1367 You probably wrote something like this:
1371 which puts literal commas into some of the list items. Write it without
1372 commas if you don't want them to appear in your data:
1376 =item Scalar value @%s{%s} better written as $%s{%s}
1378 (W) You've used a hash slice (indicated by @) to select a single element of
1379 a hash. Generally it's better to ask for a scalar value (indicated by $).
1380 The difference is that C<$foo{&bar}> always behaves like a scalar, both when
1381 assigning to it and when evaluating its argument, while C<@foo{&bar}> behaves
1382 like a list when you assign to it, and provides a list context to its
1383 subscript, which can do weird things if you're expecting only one subscript.
1385 =item Stub found while resolving method `%s' overloading `%s' in package `%s'
1387 (P) Overloading resolution over @ISA tree may be broken by importing stubs.
1388 Stubs should never be implicitely created, but explicit calls to C<can>
1391 =item Too late for "B<-T>" option
1393 (X) The #! line (or local equivalent) in a Perl script contains the
1394 B<-T> option, but Perl was not invoked with B<-T> in its argument
1395 list. This is an error because, by the time Perl discovers a B<-T> in
1396 a script, it's too late to properly taint everything from the
1397 environment. So Perl gives up.
1399 =item untie attempted while %d inner references still exist
1401 (W) A copy of the object returned from C<tie> (or C<tied>) was still
1402 valid when C<untie> was called.
1404 =item Unrecognized character %s
1406 (F) The Perl parser has no idea what to do with the specified character
1407 in your Perl script (or eval). Perhaps you tried to run a compressed
1408 script, a binary program, or a directory as a Perl program.
1410 =item Unsupported function fork
1412 (F) Your version of executable does not support forking.
1414 Note that under some systems, like OS/2, there may be different flavors of
1415 Perl executables, some of which may support fork, some not. Try changing
1416 the name you call Perl by to C<perl_>, C<perl__>, and so on.
1418 =item Use of "$$<digit>" to mean "${$}<digit>" is deprecated
1420 (D) Perl versions before 5.004 misinterpreted any type marker followed
1421 by "$" and a digit. For example, "$$0" was incorrectly taken to mean
1422 "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004.
1424 However, the developers of Perl 5.004 could not fix this bug completely,
1425 because at least two widely-used modules depend on the old meaning of
1426 "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the
1427 old (broken) way inside strings; but it generates this message as a
1428 warning. And in Perl 5.005, this special treatment will cease.
1430 =item Value of %s can be "0"; test with defined()
1432 (W) In a conditional expression, you used <HANDLE>, <*> (glob), C<each()>,
1433 or C<readdir()> as a boolean value. Each of these constructs can return a
1434 value of "0"; that would make the conditional expression false, which is
1435 probably not what you intended. When using these constructs in conditional
1436 expressions, test their values with the C<defined> operator.
1438 =item Variable "%s" may be unavailable
1440 (W) An inner (nested) I<anonymous> subroutine is inside a I<named>
1441 subroutine, and outside that is another subroutine; and the anonymous
1442 (innermost) subroutine is referencing a lexical variable defined in
1443 the outermost subroutine. For example:
1445 sub outermost { my $a; sub middle { sub { $a } } }
1447 If the anonymous subroutine is called or referenced (directly or
1448 indirectly) from the outermost subroutine, it will share the variable
1449 as you would expect. But if the anonymous subroutine is called or
1450 referenced when the outermost subroutine is not active, it will see
1451 the value of the shared variable as it was before and during the
1452 *first* call to the outermost subroutine, which is probably not what
1455 In these circumstances, it is usually best to make the middle
1456 subroutine anonymous, using the C<sub {}> syntax. Perl has specific
1457 support for shared variables in nested anonymous subroutines; a named
1458 subroutine in between interferes with this feature.
1460 =item Variable "%s" will not stay shared
1462 (W) An inner (nested) I<named> subroutine is referencing a lexical
1463 variable defined in an outer subroutine.
1465 When the inner subroutine is called, it will probably see the value of
1466 the outer subroutine's variable as it was before and during the
1467 *first* call to the outer subroutine; in this case, after the first
1468 call to the outer subroutine is complete, the inner and outer
1469 subroutines will no longer share a common value for the variable. In
1470 other words, the variable will no longer be shared.
1472 Furthermore, if the outer subroutine is anonymous and references a
1473 lexical variable outside itself, then the outer and inner subroutines
1474 will I<never> share the given variable.
1476 This problem can usually be solved by making the inner subroutine
1477 anonymous, using the C<sub {}> syntax. When inner anonymous subs that
1478 reference variables in outer subroutines are called or referenced,
1479 they are automatically rebound to the current values of such
1482 =item Warning: something's wrong
1484 (W) You passed warn() an empty string (the equivalent of C<warn "">) or
1485 you called it with no args and C<$_> was empty.
1487 =item Ill-formed logical name |%s| in prime_env_iter
1489 (W) A warning peculiar to VMS. A logical name was encountered when preparing
1490 to iterate over %ENV which violates the syntactic rules governing logical
1491 names. Since it cannot be translated normally, it is skipped, and will not
1492 appear in %ENV. This may be a benign occurrence, as some software packages
1493 might directly modify logical name tables and introduce nonstandard names,
1494 or it may indicate that a logical name table has been corrupted.
1496 =item Got an error from DosAllocMem
1498 (P) An error peculiar to OS/2. Most probably you're using an obsolete
1499 version of Perl, and this should not happen anyway.
1501 =item Malformed PERLLIB_PREFIX
1503 (F) An error peculiar to OS/2. PERLLIB_PREFIX should be of the form
1511 with nonempty prefix1 and prefix2. If C<prefix1> is indeed a prefix
1512 of a builtin library search path, prefix2 is substituted. The error
1513 may appear if components are not found, or are too long. See
1514 "PERLLIB_PREFIX" in F<README.os2>.
1516 =item PERL_SH_DIR too long
1518 (F) An error peculiar to OS/2. PERL_SH_DIR is the directory to find the
1519 C<sh>-shell in. See "PERL_SH_DIR" in F<README.os2>.
1521 =item Process terminated by SIG%s
1523 (W) This is a standard message issued by OS/2 applications, while *nix
1524 applications die in silence. It is considered a feature of the OS/2
1525 port. One can easily disable this by appropriate sighandlers, see
1526 L<perlipc/"Signals">. See also "Process terminated by SIGTERM/SIGINT"
1533 If you find what you think is a bug, you might check the headers of
1534 recently posted articles in the comp.lang.perl.misc newsgroup.
1535 There may also be information at http://www.perl.com/perl/, the Perl
1538 If you believe you have an unreported bug, please run the B<perlbug>
1539 program included with your release. Make sure you trim your bug down
1540 to a tiny but sufficient test case. Your bug report, along with the
1541 output of C<perl -V>, will be sent off to <F<perlbug@perl.com>> to be
1542 analysed by the Perl porting team.
1546 The F<Changes> file for exhaustive details on what changed.
1548 The F<INSTALL> file for how to build Perl. This file has been
1549 significantly updated for 5.004, so even veteran users should
1552 The F<README> file for general stuff.
1554 The F<Copying> file for copyright information.
1558 Constructed by Tom Christiansen, grabbing material with permission
1559 from innumerable contributors, with kibitzing by more than a few Perl
1562 Last update: Wed May 14 11:14:09 EDT 1997